A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
The H3D Dataset for Full-Surround 3D Multi-Object Detection and Tracking in Crowded Urban Scenes
[article]
2019
arXiv
pre-print
Preserved Fulltext
3D multi-object detection and tracking are crucial for traffic scene understanding. ...
With unique dataset size, rich annotations, and complex scenes, H3D is gathered to stimulate research on full-surround 3D multi-object detection and tracking. ...
Acknowledgement: We are grateful to our colleagues Behzad Dariush, Kalyani Polagani, Kenji Nakai, Athma Narayanan, and Wei Zhan for their valuable input. ...
arXiv:1903.01568v1
fatcat:7hvhnqsb6fbktprsmuirhh2jha
3D-SiamRPN: An End-to-end Learning Method for Real-time 3D Single Object Tracking using Raw Point Cloud
2020
IEEE Sensors Journal
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2108.05630v1 fatcat:vgg56b35kvganna3ryim5e6ewq
3D single object tracking is a key issue for autonomous following robot, where the robot should robustly track and accurately localize the target for efficient following. ...
Additionally, experimental results on H3D dataset demonstrate that our method also has good generalization ability and could achieve good tracking performance in a new scene without re-training. ...
ACKNOWLEDGMENT The authors would also like to thank J. Shan, W. Qiao and M. Zhou for their help. ...
doi:10.1109/jsen.2020.3033034
fatcat:uqd5gowmlrci5nrqwusbyn6s6i
IPS300+: a Challenging Multimodal Dataset for Intersection Perception System
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Due to the high complexity and occlusion, insufficient perception in the crowded urban intersection can be a serious safety risk for both human drivers and autonomous algorithms, whereas CVIS (Cooperative ...
The first batch of open-source data includes 14198 frames, and each frame has an average of 319.84 labels, which is 9.6 times larger than the most crowded dataset (H3D dataset in 2019) by now. ...
After the ID checks for targets are completed, multiple ID documents in 5Hz, 20s fragments of the published data will be released for 3D multi-target tracking task. ...
arXiv:2106.02781v1
fatcat:wsemjp6tbrbhfi7nhwnt2fr7wm
nuScenes: A multimodal dataset for autonomous driving
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Image based benchmark datasets have driven development in computer vision tasks such as object detection, tracking and segmentation of agents in the environment. ...
Robust detection and tracking of objects is crucial for the deployment of autonomous vehicle technology. ...
The nuScenes dataset was annotated by Scale.ai and we thank Alexandr Wang and Dave Morse for their support. ...
arXiv:1903.11027v5
fatcat:ha265rjm4bbndnnugq5ususime
Multi-Modal 3D Object Detection in Autonomous Driving: a Survey
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Next, we discuss some popular datasets for multi-modal 3D object detection, with a special focus on the sensor data included in each dataset. ...
We hope that our detailed review can help researchers to embark investigations in the area of multi-modal 3D object detection. ...
It first calculates mAP of each class and
then averaging over the 3 classes as the final detec-
tion result.
-H3D (Patil et al., 2019) focuses on crowded traffic
scenes in urban. ...
arXiv:2106.12735v2
fatcat:5twzbk4yhrcfzddp7zghnsivna
TITAN: Future Forecast using Action Priors
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In the absence of an appropriate dataset for this task, we created the TITAN dataset that consists of 700 labeled video-clips (with odometry) captured from a moving vehicle on highly interactive urban ...
traffic scenes in Tokyo. ...
Acknowledgement We thank Akira Kanehara for supporting our data collection and Yuji Yasui, Rei Sakai, and Isht Dwivedi for insightful discussions. ...
arXiv:2003.13886v3
fatcat:djx7cu6blrddlahtodtekom3am
Mid-level Representation for Visual Recognition
[article]
2015
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In the case of image understanding, we focus on object detection/recognition task. ...
We, additionally, study the outcomes provided by employing the subcategory-based models for undoing dataset bias. ...
For instance, Zhao and Nevatia [162] used 3D human models to detect persons
in the observed scene as well as a probabilistic framework for tracking extracted
features from the persons. ...
arXiv:1512.07314v1
fatcat:knmhkwxqk5aczis7ce6g2sv2wm